Haykin Kalman Filtering and Neural Networks

1. Auflage 2004
ISBN: 978-0-471-46421-1
Verlag: John Wiley & Sons
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 304 Seiten, E-Book

Reihe: Adaptive and Cognitive Dynamic Systems: Signal Processing, Learning, Communications and Control

State-of-the-art coverage of Kalman filter methods for thedesign of neural networks
This self-contained book consists of seven chapters by expertcontributors that discuss Kalman filtering as applied to thetraining and use of neural networks. Although the traditionalapproach to the subject is almost always linear, this bookrecognizes and deals with the fact that real problems are mostoften nonlinear.
The first chapter offers an introductory treatment of Kalmanfilters with an emphasis on basic Kalman filter theory,Rauch-Tung-Striebel smoother, and the extended Kalman filter. Otherchapters cover:
* An algorithm for the training of feedforward and recurrentmultilayered perceptrons, based on the decoupled extended Kalmanfilter (DEKF)
* Applications of the DEKF learning algorithm to the study ofimage sequences and the dynamic reconstruction of chaoticprocesses
* The dual estimation problem
* Stochastic nonlinear dynamics: the expectation-maximization(EM) algorithm and the extended Kalman smoothing (EKS)algorithm
* The unscented Kalman filter
Each chapter, with the exception of the introduction, includesillustrative applications of the learning algorithms describedhere, some of which involve the use of simulated and real-lifedata. Kalman Filtering and Neural Networks serves as an expertresource for researchers in neural networks and nonlinear dynamicalsystems.
An Instructor's Manual presenting detailed solutions to all theproblems in the book is available upon request from the WileyMakerting Department.

Autoren/Hrsg.

Haykin, Simon

Inhaltsverzeichnis

Preface.
Contributors.
Kalman Filters (S. Haykin).
Parameter-Based Kalman Filter Training: Theory and Implementaion(G. Puskorius and L. Feldkamp).
Learning Shape and Motion from Image Sequences (G. Patel, etal.).
Chaotic Dynamics (G. Patel and S. Haykin).
Dual Extended Kalman Filter Methods (E. Wan and A. Nelson).
Learning Nonlinear Dynamical System Using theExpectation-Maximization Algorithm (S. Roweis and Z.Ghahramani).
The Unscencted Kalman Filter (E. Wan and R. van der Merwe).
Index.

Über Autor(innen)

SIMON HAYKIN, PhD, is Professor of Electrical Engineering at the Communication Research Laboratory of McMaster University in Hamilton, Ontario, Canada.